Raid Weirdness -- /dev/md0: Invalid argument
Hello all,
This was all working just fine and dandy... Then this guy named murphy came by, and he told me that if something goes wrong you won't have a clue what happened or how it happened. And Damnit he was right. The bearings on my power supply fan went bad and the noise was driving me mad, so I halted my fileserver and put in a new power supply (I didn't have any spare fans laying around, but I did have a new power supply laying around). When I turned my fileserver back on the Raid didn't come up. When I turned my fileserver back on the Raid didn't come up. Running: raidstart /dev/md0 yeilded: md: array md0 already exists! So I: raidstop /dev/md0 raidstart /dev/md0 and I got: /dev/md0: Invalid argument looking in dmesg it showed more detail: [events: 00000000] md: invalid raid superblock magic on ide/host2/bus0/target0/lun0/part1 md: ide/host2/bus0/target0/lun0/part1 has invalid sb, not importing! md: could not import ide/host2/bus0/target0/lun0/part1! md: autostart ide/host2/bus0/target0/lun0/part1 failed! I did an lsmod to verify that the appropriate raid drivers were loaded, and they were. The disks are fine, both show up in the bios search and under /proc. cat'ing the drives produced copious amounts of binary gobbly gook which is nice. I know my /etc/raidtab is fine (I will include it anyway) because this has been working for the past month since I first built this raid array. Any suggestions/pointers on how to recover this raid array without loss of data? here is the relevent parts of my /etc/raidtab and /etc/fstab ### RaidTab ### raiddev /dev/md0 raid-level 0 nr-raid-disks 2 persistent-superblock 1 chunk-size 4 device /dev/hde1 raid-disk 0 device /dev/hdg1 raid-disk 1 ### FSTab ### /dev/md0 /Raid reiserfs rw,user,exec 0 0 Other Misc info: Debian Sarge. Home Grown 2.4.24 kernel. Raid is on 2 160GB drives on a promise tx2 Ultra100 ide PCI controller card. Cheers, Silas Bennett =0) |
Followup:
Hello again, I stuck a Knoppix (3.3) disk in to see if I could get it to work. Well it didn't work... But I got a bit more usefull info out of dmesg using knoppix. <Snip from Previous Post> [events: 00000000] md: invalid raid superblock magic on ide/host2/bus0/target0/lun0/part1 md: ide/host2/bus0/target0/lun0/part1 has invalid sb, not importing! md: could not import ide/host2/bus0/target0/lun0/part1! md: autostart ide/host2/bus0/target0/lun0/part1 failed! </Snip from Previous Post> really means: [events: 00000000] md: invalid raid superblock magic on hde1 md: hde1 has invalid sb, not importing! md: could not import hde1! md: autostart hde1 failed! I still would like to get this raid back up without repartitioning/rebuilding-raid/reformating and loosing all of my data in the process. Any help, insight, or RTFMs follow by links to a helpfull manual would be greatly appreciated. Cheers, Silas Bennett =0) |
I wouldnt doubt this is realtated to an improper/poor voltage output of your power supply.
I would buy a brand new power supply rated for your mobo and hook it up. |
Hi,
Thanks for the reply. Quote:
Any other Suggestions/Insights/RTFMs? Cheers, Silas Bennett =0) |
All times are GMT -5. The time now is 09:49 AM. |